In addition to have JupyterLab installed, you will need to the following Python packages to run this notebook.
(The Jupyter Notebook that generated this page is available here.)
The ground-based University of Wisconsin (UW)-Madison scanning High Spectral Resolution Lidar (HSRL) L1B (similar to a NASA CALIOP L1B product) scanning product consists of the following geophysical 532nm wavelength calibrated measurements:
The calibrated 1064nm attenuated backscatter [$\mathrm{m}^{-1}\,\mathrm{sr}^{-1}$] will be added in the future, along with uncertainties (i.e. standard deviation) of each geophysical variable.
The UW HSRL produces these measurements in three viewing angle configurations.
These three viewing measurement configurations are organized in their own netcdf4 groups which will be discussed in the subsequent subsections. Here is how you download the sample (prelimary) 1.4 GB L1B netcdf4 file.
The L1B file name consists of the following parts:
<instrument name>_<start date>_<end date>_<time resolution>s_<altitude resolution>m_<angle resolution>deg_L1B<_tag>.nc
For example the specific UW HSRL instrument is designated as the bagohsrl. The file consists of measurements of the 4th of August where the nominal time and altitude resolutions are 30 seconds and 7.5 meters, and the scanning data has a (downsampled) angular resolution of 1 degree. Finally the tag preliminary provides a label of the file.
Similar to the CALIOP L1B product the geophysical variable names encode the wavelength. The self explanatory variable names are
particulate_backscatter_532nmparticulate_linear_depolarization_532nmparticulate_optical_depth_532nmattenuated_color_ratio_1064nm_532nmHere is how you get the vertical stare data via xarray. The HSRL geophysical variable dimensions are time and mean sea level (MSL) altitude. When you hover your mouse cursor over the variable name under Data variables you will see the full name. The document icon button to the far right, when clicked upon, shows you the attributes of the corresponding variable.
<xarray.Dataset>
Dimensions: (time: 361, altitude: 2668)
Coordinates:
* time (time) datetime64[ns] 2023-08-14...
* altitude (altitude) float64 7.223 ... 1.9...
Data variables:
particulate_linear_depolarization_532nm (time, altitude) float64 ...
optical_depth_532nm (time, altitude) float64 ...
particulate_optical_depth_532nm (time, altitude) float64 ...
particulate_backscatter_532nm (time, altitude) float64 ...
attenuated_color_ratio_1064nm_532nm (time, altitude) float64 ...
Attributes:
documentation: github.com/ssec/hsrlHere are the vertical stare geophysical variables plots. The white columns in the images are the time intervals when the HSRL was either making horizontal stares or scanning. The rest of the white areas, above 5km altitude, are low signal to noise ratio (SNR) measurements that were masked out. Note that the particulate optical depth has negative values, since we are in the process of post calibrating the HSRL.
OMP: Info #276: omp_set_nested routine deprecated, please use omp_set_max_active_levels instead.
Unlike the vertical stare measurements, the dimensions of the horizontal stare geophysical variable dimensions are time and range, i.e. the optical path distance between the instrument and the atmospheric volume. Within the netcdf4 group two dimensional altitude coordinates are provided, as a function of time.
(Note: The horizontal stare viewing angle will be added in the near future.)
Here are the horizontal stare geophysical variables plots. The white columns in the images are the time intervals when the HSRL was either making vertical stares or scanning. The rest of the white areas, above 5km altitude, are low signal to noise ratio (SNR) measurements that were masked out. Note that the particulate optical depth has negative values, since we are in the process of post calibrating the HSRL.
The following plot shows the altitude of the measurements at ranges 5km.
The scanning measurement dimensions are angle (i.e. scan angle), range and the scan start time by which each individual scan is indexed. Accompanied with these dimensions are the corresponding altitude and distance coordinates which can be used in plotting routines.
In the following plots low SNR measurements are masked out.
Here is how you make a timeseries plot of the scanning measurements over altitude and distance intervals $(60\mathrm{m}, 100\mathrm{m})$ and $(2980\mathrm{m}, 3020\mathrm{m})$.